multimoded rf systems for future linear...

Multimoded RF Systems for Future Linear Colliders

Sami G. Tantawi

AcknowledgmentThis work is a result of a continuous effort by many researches and engineers over many years.

In particular, The efforts of C. Nantista, N. Kroll, R. Miller, P. Wilson, V. Dolgashev, K. Fant, C. Pearson, R. Ruth, were instrumental to the results achieved to date.

Jose Chan: A persistence efforts during the manufacturing and testing of components

David Schultz Denis Atkinson, C. Adolphsen D. Burke and the NLC group

J. Nelson, K. Jobe, M. Ross, J. Frisch, T. Smith, D. McCormick, and the NLCTA operations team were a must for the successful conclusion of this project.

D. Cassel, J. DeLamare, M. Nguyen, M. Laruss and the modulator team

K. Ratcliffe and the Vacuum Assembly Team.

We wish to Thank all operators that took shifts 24 hours a day for several months

Outline

• Dualmode resonant delay line pulse compression system for the Next Linear Collider (NLC)

1. Introduction2. Components: design and cold tests3. Dualmode Delay Lines: Design and

Experimental results4. High power experimental results

RF Accelerator Structures needs a short, high power, rf

pulse. For example the current NLC design requires a flat-top

396 ns rf pulse with a power level of about 95 MW/m at 11.424 GHz, and

a repetition rate of 120 Hz to feed each accelerator structure.

One need to transfer the CW wall plug power to rf pulses with high

power and low duty factor. Hence

1. A storage system is needed.

2. A Switch or a switching mechanism is needed to control the charging and

discharging of the system.

3. A device to generate the RF power

Storage Systems:1. Capacitors; i.e. Modulators.

Switches include pulse forming networks, thyrotrons, IGBTs, and grided guns on microwave tubes.

2. Kinetic energy of an electron beam ; i.e. two beam accelerator, Switches include RF beam kickers

3. In the accelerator structure; i.e. super conducting accelerators.

4. In rf transmission lines and cavities; i.e. rf pulse compression systems. Switches includes rf phase manipulation between rf sources, andsolid state switches.

In general most of rf systems, suggested for a linear collider, contains elements from several of the above storage system.

To compare these system one has to consider: 1- Philosophy of the design

• Modularity: one may choose to have a unit that contains one rf source, and compact pulse compression system such as SLED-II• Flexibility in the operating rf frequency: one may choose Two beam systems•…..

2- Efficiency3- Cost

The choice of the system is greatly affected by the available technology of the components.

If one has

1- A really inexpensive efficient rf source

2- A a very efficient and inexpensive modulator system using a very fast switch (such as grided gun on the rf source)

One might use several thousands of these devices to power the main Linac of a collider

However, neither the inexpensive rf source nor the very fastmodulator exist, only ideas at the moment.

The need for RF Pulse compression

1- It is usually easier to build rf sources with low power and long pulse width. 2- Rf sources are expensive, one should get as much energy from them as possible, i.e., the longest possible pulse. One should not let an expensive rf source in an idle mode most of the time.

Hence, rf pulse compression is needed to match the long pulse low power of available rf sources such as klystrons to the high power short pulse needed for the accelerator structure.

• Pulse Compression should be used with as high of a compression ratio as possible until

a) The cost of the compression system starts to exceed that of the klystrons and modulator

b) Or, the available pulse width of the klystrons is exhausted

• The cost of the main linac is directly proportional to the efficiency of the pulse compression system.

RF Pulse compression for RF Linacs and Colliders

• Resonance Delay Lines (SLED-II)• Binary Pulse Compression (BPC)• Delay Line Distribution System (DLDS)

• Two-Stage systems, any combination of two or more of the above systems

klystrons

Single or Multi-Moded Delay Lines

Short Circuit

Coupling Irises

Accelerator Structures

3 dB 90 Degree Hybrid

Sled-II Pulse compression system

klystrons


Short Circuit

Coupling Irises (can be actively switched)

Circulator


Sled-II pulse compression system with a circulator and active switches

Single-Moded Delay Lines


Accelerator Structure

Two banks of power sources each has an nk/2 klystrons

a) Single-moded Binary Pulse Compression


Accelerator Structure

Two banks of power sources each has an nk/2 klystrons


Circulator

Short Circuit

b) Binary pulse compression can have several improvements including the use of a circulator and several modes to reduce the delay line length.

Delay Lines

Accelerator StructuresBank of nk of klystrons

A set of hybrids that switches the combined rf to different outputs

Not all the output need to be used. The unused outputs are terminated by an rf load

a) A Unit of a Single-Moded DLDS

Multi-Moded Delay Lines. The total number of these lines is np

A mode launcher which takes nm inputs and produces nm modes into a single waveguide delay line

b) A Unit of a Multi-Moded DLDS

Single-Moded Delay Lines


A combinerA High Power Microwave Switch

c) A Unit of an Active DLDS

The challenges facing most of these pulse compression system are

1- Compactness, how to produce a storage system which is relatively compact

2- Efficiency, for the resonant delay lines, efficiency could be boosted by an rf switch.

3- Most of these systems could have a more compact topology if one have, a nonreciprocal RF device (circulator), or a switch.

Compactness

•A waveguide near cutoff, hence, a low group velocity. One can use a higher order mode to reduce the losses. A bad idea, dispersion will destroy the pulse shape at a group velocity less than 4.

•Loaded waveguide (slow wave structure). A bad idea because of dispersion and losses.

•Multimoded Waveguide, the only good idea with no draw backs. We are using highly overmoded waveguide systems no mater what, using an extra mode, or two, or four, .. Is a bonus.

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

4 6 8 10 12 14 16

Single Moded DLDSMulti-Moded DLDS (number of modes=3)Active DLDSMulti-Moded BPC (A high power circulator and 3 modes)Multi-Moded SLED II (A high power circulator and 3 modes) Active SLED II (One time Switching [7])Multi-Moded DLDS (n

k=4, number of modes =3)

Single-Moded DLDS (nk=4)

Rel

ativ

e C

ost

Compression Ratio

nk=8

0

20

40

60

80

100

0 200 400 600 800 1000 1200 1400

Ele

ctric

Fie

ld (M

V/m

)

Pulse Width [ns]

Electric Field breakdown strength in a 16% Group velocity copper waveguide at 11.424 GHz

Pulsed Heating

0

10

20

30

40

50

60

70

0 50 100 150 200 250 300 350

400ns800ns1600ns3200ns

Pul

sed

Hea

ting

Sur

face

Tem

pera

ture

Ris

e (°

C)

|Hs| (kA/m)

Pulsed Heating of Copper Surface Vs. Surface H Field for Different Pulse Durations

Dualmode Resonant Delay lines ~30m

RF Input to the 4 50 MW klystrons

Single mode waveguide input to the pulse compression system; 100 MW/Line for 1.6 µs

Dual mode waveguide carrying 200 MW

Compressed output > 600 MW 400 ns.

Output Load Tree

NLC experimental rfpulse compression

system

Dualmode CombinerTwo fundamental

mode inputs

TE10/TE20

TE01/TE11

Dualmode directional coupler

Pumpout

Super Hybrid

TE01 to delay Lines

TE01/TE11 to loads

The Head of the pulse compression system

139.8 mm

40.64 mm

36.45 mm

30.58 mm

WC160

Section#1

Section#2

Section#3

Width taper to match our standard overmoded waveguide

36.63 mm

TE01

TE20

TE10

TE11

TE01

TE20

TE10

TE11

Simulated electric fields (HFSS)of the multi-moded circular to rectangular taper

139.8 mm

Taper Geometry (Operating Frequency=11.424 GHz)

40.64 mm

36.45 mm 36.63 mm

Dual-Moded Rectangular Circular Converter/Tapers

TE01

TE20

TE10

TE11

DualmodeRectangular-to-Circular

Taper

Instrumental components for cold testing of multimode components:

1. TE11 Mode launcher2. TE01 Mode launcher3. Width taper4. Height taper5. Small waveguide sections with different lengths at all

waveguide cross sections

We followed a strict methodology of designing these instruments. They had to be simulated with at least three different codes and have a performance that is much better than any component that we have. Of course we can only do that because there is no restrictions on field levels.

Cold Test Components/Calibration Standards

Width Taper, Height Tapers and Jog Mode Converter

TE01 Mode Converter, TE11 Mode Converter, and Size Taper

Multimoded Matched Load

11.324 11.357 11.391 11.424 11.457 11.491 11.5240.9

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1

before modificationafter modification

Frequency [GHz]

S12

Measured S12 between the Rectangular TE02 mode and the circular TE01mode. These measurements include the response of mode transducers

necessary to launch the modes at both ends of the taper.

0.9

0.92

0.94

0.96

0.98

1

11.324 11.374 11.424 11.474 11.524

S12

Frequency (GHz)

Measured S12 between the rectangular TE01 mode and the circular TE11 mode. These measurements include the response of mode transducers necessary to launch the modes at

both ends of the taper.

To instrumental width taper-To instrumental height taper-To Network analyzer

To instrumental width taper-To instrumental height taper-To Network analyzer

TE01 Mode launcher

From Network Analyzer

Cold Test Setup for the Splitter and Circular-to-Rectangular Taper

TE20 Mode

3

2

1

170

175

180

185

190

11.3 11.35 11.4 11.45 11.5 11.55

Phase diiference between S21

and S31

Phas

e (d

egre

es)

Frequency (GHz)

-4

-3.8

-3.6

-3.4

-3.2

-3

11.274 11.324 11.374 11.424 11.474 11.524 11.574

S21

S31

Mag

nitu

ed (d

B)

Frequency (GHz)

Cold Test Results of:Splitter-Circular to rectangular taper-Wraparound mode converter-instrumental height taper-instrumental width taper. Total losses at 11.424 GHz =1.3%

Jog Converter and Mode Mixer

1.442”TE10 TE10

orthogonal 50/50 mixesTE20

TE20

Magic H Hybrid

0.900”

2.3821”

1.4788”

@ 600 MW, 1.435” height: |Esmax| = ~45.6 MV/m

|Hsmax| = ~168 kA/m

C. Nantista ‘02

TE10

straight through

TE02

To the pulse compressor

@ 516 MW,

|Esmax| = ~45.8 MV/m

|Hsmax| = ~156 kA/m To/From Pulse

Compressors

Jog-ConverterCombiner/Splitter

Planer 3-dB Hybrid

Jog mode mixer

Combiner/Splitter

Planer 3-dB Hybrid

Planer Super-Hybrid

Port 1

Port 4

Port 3

Port 2

Circular to Rectangular Tapers

Jog-mode converter

Mitered Bends

-0.5

-0.4

-0.3

-0.2

-0.1

0

11.404 11.414 11.424 11.434 11.444

ports 2&3 shortedport 2 openports 2&3 openTE

11 T

rans

mis

sion

(dB

)

Frequency (GHz)

Circular TE11 mode transmission and reflection between port 1 and port 4 (S41) for different

conditions at port 2 and port 3.

Port 1

Port 4Port 3

Port 2

-60

-50

-40

-30

-20

-10

0

11.324 11.374 11.424 11.474 11.524

ports 2&3 shortedport 2 openports 2&3 open

TE11

Ref

lect

ion

(dB

)

Frequency (GHz)

-60

-50

-40

-30

-20

-10

0

11.404 11.414 11.424 11.434 11.444

initialspacer added to input

TE01

Ref

lect

ion

(dB)

Frequency (GHz)

Port 1

Port 4Port 3

Port 2

Reflection measurements (S11) for the circular TE01 mode from port 1 while shorting both port 2 and port 3 and matching port 4.

Port 1

Port 4Port 3

Port 2

-0.5

-0.45

-0.4

-0.35

-0.3

-0.25

-0.2

-0.15

-0.1

11.404 11.414 11.424 11.434 11.444

TE01

Tra

nsm

issi

on (d

B)

Frequency (GHz)

Circular TE01 mode transmission between port 1 and port 4 (S41) while shorting both port 2 and port 3.

1.442”

@ 600 MW |Esmax| = ~37.8 MV/m


Bend Converter Design

Dual-Mode Splitter

TE01 or TE11

TE10

TE01

TE20bend converter

Dualmode Splitter: For either incident mode the power is evenly divided between the two output ports, which launch the TE01.

Load Tree: The input power, carried by the TE01 mode, is split 4 ways to be absorbed at the loads

WC160

|S11| < -57 dB @ 11.424 GHz

Four Way Splitter DesignWR90

-7

-6.5

-6

-5.5

11.324 11.374 11.424 11.474 11.524

Power Divider

Port 1Port 2

Port 3Port 4

dB

Frequency (GHz)

-35

-30

-25

-20

-15

-10

11.324 11.374 11.424 11.474 11.524

Circular Port

dB

Frequency (GHz)

High Power Load Design

Rectangular waveguide for coupling the TE01 mode

Ridge waveguide for coupling the TE11 mode

Circular Waveguide

•The waveguide sizes are chosen to match wavelengths between the circular waveguide modes and side waveguide fundamental mode

•The coupling hole pattern represents a Hamming window

Dual-moded Directional Coupler

End taper for TE11 coupler

End taper for the TE01 coupler

Directional Coupler Cold Test

-125

-120

-115

-110

-105

-100

-95

-90

-85

11.174 11.299 11.424 11.549 11.674

Coupling between the TE01

mode and the TE11

coupler arm

dB

Frequency (GHz)

Since the coupling coefficient for the desired mode is -47 dB, Isolation between coupler arm and the unwanted mode is better than -45 dB.

Measured directivity for the TE01 Arm

Pumpout design: the set of holes are designed to cancel any coupling orself-coupling for the TE01 and the TE11

Vacuum Pumpout Cold Test

0

0.05

0.1

0.15

0.2

0.25

0.3

11.35 11.4 11.45 11.5

WC159 Vacuum Pumpout Match

TE11TE01

Am

plitu

de o

f S11

Frequency (GHz)

0.86

0.88

0.9

0.92

0.94

0.96

0.98

1

11.35 11.4 11.45 11.5

WC159 Vacuum Pumpout Transmission

TE11TE01

Frequency (GHz)

Am

plitu

de o

f S21

TE01

TE02

TE02

TE01

Dual Moded Delay line occupy only half the length of a single moded delay line

Dual-Moded Delay LineDual-moding the delay lines cuts their required length approximately in half.

21

21

2 gg

gg

vvvvTL+

=

input taper

17.08 cm delay line ~29 m long

end taper

TE01

TE02

TE02

TE01

mode converter / tuning short

R=4.1275cm

vacuum feed-through plunger

TE01(TE02)

⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−−−

−−−

=

θθθ

θθθ

θθθ

φφ

φφ

cos2

sin2

sin2

sin2cos

2cos

2sin

2cos

2cos

jj

jj

ee

ee

S

Reflective TE01/TE02 Mode Converter

TE02(TE01)

S =

⎟⎟⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜⎜⎜

⎝

⎛

−−−

−−−

=

θθθ

θθθ

θθθ

φφ

φφ

cos2

sin2

sin2

sin2cos

2cos

2sin

2cos

2cos

jj

jj

ee

ee

S

1.932” 3.25”

1.045”

Field Pattern within the device. Peak field at 300 MW (one device per transmission line) is 26.6 MV/m. This field is in the middle of the small guide.

TE01

TE02

TE02

TE01

End mode converter simulations

TE01 TE02 End taper and mode converter

-0.1

-0.08

-0.06

-0.04

-0.02

0

11.324 11.374 11.424 11.474 11.524

TE01 Transmitted TE02 Transmission

dB

Frequency [GHz]

End Taper (before the TE01-TE02 Mode converter)

Input taper design

-1

-0.8

-0.6

-0.4

-0.2

0

11.3 11.35 11.4 11.45 11.5 11.55

TE01 Transmitted TE02 Reflection

dB

Frequency [GHz]

Input Taper for a Dual-Moded System

0

0.2

0.4

0.6

0.8

1

1.2

-200 -100 0 100 200 300 400 500

Reflected PulseInput Pulse

Mag

nitu

de

Time [ns]

Measured Delay through 75 feet of WC475 waveguide terminated with a flat plate. The round trip delay time is 154 ns

0

0.2

0.4

0.6

0.8

1

1.2

-200 -100 0 100 200 300 400 500

InputPulse Mag

Mag

nitu

de

Time [ns]

Measured delay through 75 feet of WC475 waveguide terminated with a TE01-TE02 Mode converter. The round trip delay time is 320 ns

Multimoded Taper Cold Test

0

0.2

0.4

0.6

0.8

1

11.3 11.35 11.4 11.45 11.5 11.55

, p p p p

Dat

a M

ag

Freq GHz

S11 M

agnitude

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

-300 -200 -100 0 100 200 300

Output pulse (delayed by 15 ns 4 round trips)Input pulse

Pul

se M

agTime(ns)

Pulse Am

plitude

Measured frequency response and constructed time response of the dual-moded taper assembly.

Measured Response of the dual-mode SLED-II Pulse compression system at a compression ratio of 4. Delay line length is ~35 feet. Output pulse width is 150 ns.

0

0.5

1

1.5

2

2.5

3

3.5

4

0 125 250 375 500 625 750 875

Compressed Pulse Input

Rel

ativ

e Po

wer

Time (ns)

Delay Line Cold Tests

Delay Line Cold Tests

-1.5

-1

-0.5

0

0.5

1

1.5

2

-1000 -500 0 500 1000

sled9

Pulse MagInput

-1.5

-1

-0.5

0

0.5

1

1.5

2

-1000 -500 0 500 1000

p1sled2

Pulse MagInputtop line bottom line

Small mid-time-bin steps seen (mode impurity).

0

0.01

0.02

0.03

0.04

0.05

0.06

0

30

60

90

120

210

240

270

300

330

Bottom LineTop Line

Ref

lect

ion

0

0.01

0.02

0.03

0.04

0.05

0.06

-12 -8 -4 0 4 8 12

Bottom LineTop Line

1st R

ound

Trip

Ref

lect

ion

Am

plitu

de (n

o iri

s)

Mode Converter Cup Position (mm)

TE01 mode contamination

bottom line: ~-34 dB

top line: ~-26 dB

Amplitude oscillation due to interference with slight wrap-around mode launcher mismatch (~-48 dB).

Spurious reflection after first round trip (irises removed)

Amp. ∆φ/∆L ∆Lcup error TE01

TE02+δTE01TE02+δTE01

TE01

2β01

TE02

const. 2β01

line/taper error

(β01+β02)

TE01+δTE02

TE02+δTE01

TE01

TE02+δTE01+δTE01(β01+β02)(β02+β01)

TE02

const.

The wavelengths for these phase oscillations are 14.2 mm and 16.1 mm, respectively.

Measurements show a phase cycle over ~ 16 mm, indicating mode contamination predominantly from the delay lines and tapers.

Slight amplitude oscillations reveal a very small fixed mismatch in wrap-around mode launcher.

Non-periodic amplitude change can be attributed to longer range beating between a small cup error contamination and that from the lines/tapers (60.3 mm).

Problem Fixed by:

•Permutations of tapers

•Adjusting iris distance

•Choosing good resonant position for tuning plunger (3 within range of motion).

0

0.5

1

1.5

2

2.5

3

3.5

-1000 -500 0 500 1000

Pulse MagInput

Pow

er G

ain

Time (ns)

Top Line

0

0.5

1

1.5

2

2.5

3

3.5

-1000 -500 0 500 1000

Pulse MagInput

Pow

er G

ain

Time (ns)

Bottom Line

System test (1)

Input

output

0

0.5

1

1.5

2

2.5

3

-1000 -500 0 500 1000

Before Closing (gap on top line~1.5 mm)

Pow

er G

ain

Time (ns)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

-1000 -500 0 500 1000

Signal Amplitudes at the Launcher

Pulse MagPulse Mag

Pul

se M

ag

Time (ns)

0

0.4

0.8

1.2

1.6

2

2.4

2.8

3.2

-1000 -500 0 500 1000

Total System Response

Pow

er G

ain

Time (ns)

88

88.2

88.4

88.6

88.8

89

450 500 550 600 650 700 750 800 850

Phase of output pulse

Pha

se (D

egre

es)

Time (ns)

0

0.4

0.8

1.2

1.6

2

2.4

2.8

3.2

-1000 -500 0 500 1000

Measurments Throgh The TE01 Arm of the Directional Coupler

Pow

er G

ain

Time (ns)

0

0.005

0.01

0.015

0.02

-1000 -500 0 500 1000 1500 2000

Reflected PowerInput

Ref

lect

ed P

ower

Time (ns)

0

0.0005

0.001

0.0015

0.002

0.0025

-1000 -500 0 500 1000

Power Seen at the TE11 Arm

Pow

er

Time (ns)

0

0.4

0.8

1.2

1.6

2

2.4

2.8

3.2

-1000 -500 0 500 1000

Comparison between the signal seen at the TE11 Arm with the signal at the TE01 Arm

103 x Measured Power at the TE11 ArmMeasured Power at the TE01 Arm

Pow

er

Time (ns)

Input

output

System test (2)

0

0.5

1

1.5

2

2.5

3

-1 -0.5 0 0.5 1

before removing the spacer

Pulse MagPulse MagInput

Pul

se M

ag

Time

System Cold Tests

-90

-88

-86

-84

-82

-80

-78

-76

11.404 11.414 11.424 11.434 11.444

dB

Freq GHz

-7

-6

-5

-4

-3

-2

-1

0

11.404 11.414 11.424 11.434 11.444

leftcomb_nospacer-53-95avg

Dat

a M

ag

Freq GHz

spacer in

spacer removed

resonance between combiner and SLED head

16 steps 2 klystrons

High Power Experiments

8-Pack Phase 1

After pulse breakdown(Loads)

Upper Delay Line end pump

Upper Delay Line Input Pump

0

5 107

1 108

1.5 108

2 108

2.5 108

3 108

3.5 108

4 108

0 5 10-7 1 10-6 1.5 10-6 2 10-6

400MWpoint100ns

Channel 1Channel 2

Cha

nnel

1

Time

0

5 107

1 108

1.5 108

2 108

2.5 108

3 108

3.5 108

4 108

0 5 10-7 1 10-6 1.5 10-6 2 10-6

383MW216ns

BC

B

A

Zero Power

421 MW 360 ns






Output Load Tree


system

System Modifications

• Replaced a pumpout• Replaced the whole line of WR90• Cooled down the WR90 with fans• Hard wired the klystrons driver together






Output Load Tree


system

436 MW 360 ns

0

100

200

300

400

500

600

0 0.5 1 1.5 2

InputOutput

Pow

er (M

W)

Time(µs)

Dualmoded SLED-II Performance December 4, 2003 11 amTE

01-Input TE

01-Output

0

100

200

300

400

500

600

0 0.5 1 1.5 2

InputOutput

Pow

er (M

W)

Dualmoded SLED-II Performance December 6, 2003 1:00 am

TE01

-Input TE01

-Output

Time(µs)

0

100

200

300

400

500

600

0 0.5 1 1.5 2

Input

OutputPo

wer

(MW

)

Dualmoded SLED-II ProcessingTE

01-Input TE

01-Output

Time(µs)

0

100

200

300

400

500

600

0 0.5 1 1.5 2

Pow

er (M

W)

T im e (µs)

Input

O utput

0

2

4

6

8

10

250 300 350 400 450 500 550

Error=100 x (Detector M easurem ent-Calorim etric M easurem ent)/ Detector M easurem ent

Erro

r (%

)

Power (M W )---Detector

Klystron 1

Klystron 4

Klystron 3

Klystron 2

φ

φ

φ

TWT1

TWT2

119 MHzPLL to 11.424 GHz

Timing Switch

Voltage-controlled Switch

Bi-phase Modulator

IQ-Modulator

AFG 1

AFG 1

Computer Data AcquisitionFrom High Power rf measurements

Low-Level RF Architecture

Multimoded SLED-II output with feedback

0

100

200

300

400

500

600

0 0.5 1 1.5 2

InputOutput

Pow

er (M

W)

Time (µs)Sami Tantawi (1/27/2004)

0

37

74

111

148

185

-0.5 0 0.5 1 1.5 2 2.5 3

txt_output

B

176

178

180

182

184

1.7 1.75 1.8 1.85 1.9 1.95 2 2.05 2.1

txt_output

B

500

505

510

515

520

525

530

1.4 1.5 1.6 1.7 1.8 1.9

InputOutput

Pow

er (M

W)

Time (µs)Sami Tantawi (1/27/2004)

60 Hz

Full Run One Minute Power Data

~455.7 hrs. on total w/ ~272 trips*~98.75 hrs. @ 60 Hz w/ ~ 116 trips

~357 hrs. @ 30 Hz w/ ~ 156 trips

30 Hz

Two 26 hrs. gaps w/ no trips.

trip diode signals recorded

*“trips” here includes accidental and deliberate human induced interruption of operation.

SLED Breakdown Event

Unclear event

Klystron Breakdown event






Output Load Tree


system

SLED-II0.03 / hr at 30 Hz

(< 0.08 / hr required)

Klystron Tear (137)

SLED OutPower (MW)

WG and Tuning (76)

0 50 100 150 200 250 3000

100

200

300

400

500

600

Analysis Corrected for SLED Mistuning and Human Interference

60 Hz30 Hz

Hours of High Power Operation

Diode Trip Analysis (Since 2/11)Out of 211 trips in 365.65 hrs (30Hz equivalent)

• 29 -SLED or Combiner • 1 - Klystron 5• 15 - Klystrons 5&6• 18 - Klystron 6• 72 - Klystron 7• 28 - Klystrons 7&8• 38 - Klystron 8• 1 – Loads• 7 - ?

+ dig. Vac. faults

8-Pack Phase 2a 8-Pack Phase 2b

Power Distribution6 dB directional coupler

4.8 dB directional coupler 3 dB

directional coupler

hybrids

structures

Mode Stripper

TE01 power(to NLCTA)

TE11 power(to load tree)

1

2 3

Mode Stripper Cold Test

-50

-40

-30

-20

-10

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S11(TE01)S11(TE11)S22S33

Port

Mat

ches

(dB

)

Frequency (GHz)

-60

-50

-40

-30

-20

-10

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S31(TE11)S21(TE01)

Por

t Iso

latio

ns (d

B)

Frequency (GHz)

-1

-0.8

-0.6

-0.4

-0.2

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S21(TE11)S31(TE01)

Port

Cou

plin

gs (d

B)

Frequency (GHz)

-50

-40

-30

-20

-10

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S11S22S33S44

Port

Mat

ches

(dB

)

Frequency (GHz)

-50

-40

-30

-20

-10

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S41S32

Port

Isol

atio

ns (d

B)

Frequency (GHz)

-5

-4

-3

-2

-1

0

11.325 11.35 11.375 11.4 11.425 11.45 11.475 11.5 11.525

S21

S31

Por

t Cou

plin

gs (d

B)

Frequency (GHz)

3 dB Directional Coupler Cold Test

Vg/c 100ns Delay Length (cm) Losses/100ns (%)t 0.119911 359.484 2.78738

Dg Di

p

0.5 1 1.5 2 2.5 3

11.4

11.41

11.42

11.43

11.44

11.45

Dg=6.725”T=0.1”Di=3”P=8.500”

Freq

uenc

y

Phase

Vg/c 100ns Delay Length (cm) Losses/100ns (%)0.13649 409.156 2.35132t

Dg Di

p

0.5 1 1.5 2 2.5 3

11.4

11.41

11.42

11.43

11.44

11.45Dg=6.725”T=0.1”Di=3”P=11.5035”

Freq

uenc

y

Phase

The cell is not matched enough. We can match two cell or structure of several cells by placing the cells at right distance from each to other. But to get more smooth passbandis better to match each cell by iris.

GHz

11.374 11.384 11.394 11.404 11.414 11.424 11.434 11.444 11.454 11.464 11.474A

mpl

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

ReflectedTransmitted

Passband of single cell

Geometry of cell with matching irises.

Total length of SLED-II can be not more then 5m

Port 2 (rf)

Port 1 (rf)

Port A (Bias)

Garnet Material

Wrap-around Mode converter

Coaxial TE01and TE02 mode

propagate in this coaxial structure

Port 3 (rf)

Port 2 (rf)

Port 1 (rf)

Port A (Bias)

Garnet Material

Wrap-around Mode converter

Coaxial TE01and TE02 mode

propagate in this coaxial structure

Port 3 (rf)

Port B (Bias)

Port 1 (rf) Port 2 (rf)

Port A (Bias)

Garnet Material

Wrap-around Mode

converter

Coaxial TE01 mode propagate in this coaxial

structure

Port B (Bias)

Port 1 (rf) Port 2 (rf)

Port A (Bias)

Garnet Material

Wrap-around Mode

converter


structurePort 1 (rf) Port 2 (rf)

Port A (Bias)

Garnet Material

Wrap-around Mode

converter


structure

Wrap-Around Mode Converter for Tap-off, and extraction, tested to 470 MW

rg

ri

ro

Garnet Material

r

x

y

z

φ

Biasing magnetic field

rg

ri

ro

Garnet Material

r

x

y

z

φ

Biasing magnetic field

Conclusion:• We have introduce a fully dual mode rf system

• We have shown design and experimental data for over moded components that propagates two modes at the same time. These component perform all possible function found in single moded rf systems

• At the operating frequency of 11.424 GHz, the peak electric field is ~49 MV/m (400 ns) and the peak Magnetic field is ~0.17 MA/m (400 ns). This was demonstrated to be low enough for a reliable high power operation of the system.

Conclusion:• We have introduce a fully dual mode rf system

• We have shown design and experimental data for over moded components that propagates two modes at the same time. These component perform all possible function found in single moded rf systems

• At the operating frequency of 11.424 GHz, the peak electric field is ~49 MV/m (400 ns) and the peak Magnetic field is ~0.17 MA/m (400 ns). This should be low enough for a reliable high power operation of the system (remain to be seen)

• We have invented several new measurement techniques and instrumental components needed for characterizing dual moded rf systems.

multimoded rf systems for future linear...

Documents